An important explanation for limited global health program effectiveness is behavioral obstacles that drive a wedge between the laboratory efficacy of health technologies and their effectiveness in real-world programs. Innovative new forms of contracting that address these behavioral obstacles by rewarding both providers (on the supply-side) and consumers (on the demand-side) for ultimate social objectives like good health outcomes may therefore produce large health gains in developing countries. There is a paucity of rigorous evidence on their actual benefits in practice, however. The proposed project will produce several important advances over previous studies of pay-for- performance incentives. First, existing evidence from the handful of good-quality studies provides limited insight into the full promise of pay-for-performance incentives because they reward the use of specific health inputs rather than actual health improvement. Rewarding health outcomes directly (as we propose to do) provides stronger incentives for providers and consumers to use local information to develop innovative new health improvement strategies without rigidly prescribing about how they should do so. Second, although knowledge about the impact of demand-side incentives has recently grown, there is no evidence on their comparative effectiveness (and cost-effectiveness) relative to supply-side incentives. Third, whether or not demand- and supply-side incentives are net substitutes or complements - a key question for policymakers combining different interventions into large-scale health improvement strategies - is completely unknown. To provide new evidence on how better to align supply-side incentives with good population health and to assess their comparative effectiveness and potential synergies with demand-side incentives, we propose to conduct a large-scale randomized policy experiment in rural China. Our specific focus is the provision of incentives for reducing anemia to primary school principals and the parents of their students. Our project is the first of which we are aware that directly rewards health outcomes, and it holds the potential for immediate impact given current interest from China's government. In doing so, we propose six specific aims: (1) To provide new estimates of anemia prevalence among primary school age children in rural China; (2) To analyze the behavioral responses of educators (principals) in poor rural primary schools in China to pay-for-performance rewards for reducing anemia - and their net health consequences; (3) To estimate how private health investments that parents make in their children respond to public health interventions; (4) To study the comparative effectiveness of supply- and demand-side incentives for anemia reduction and to assess whether or not they are net complements or substitutes; (5) To calculate the cost-effectiveness of supply- and demand-side approaches for anemia reduction; and (6) To create a new, publicly available dataset of rural Chinese children and their schools.